基于视觉的运动汽车跟踪技术研究
|
摘要
基于视觉的目标跟踪技术是机器视觉的重要应用领域,已经成为当今智能交通领域的热点研究课题之一,也是智能交通监控应用的一个重要研究方向。通过对运动汽车的跟踪实现交通参数的提取以及交通事件自动检测等,而在跟踪车辆的过程中,经常会发生车辆间的相互遮挡、光线变化、复杂背景的影响以及汽车运动的随机性等问题,这些问题直接影响车辆的跟踪精度。如何实现在复杂背景、车辆遮挡、光线变化、目标机动等情况下对车辆实时稳定的跟踪是本文研究的重点。
针对汽车运动过程的机动特点,建立了机动目标的“当前”统计模型,大量文献证明“当前”统计模型在跟踪机动目标时效果良好,但是其对匀速运动或弱机动目标跟踪时性能变差。针对这一问题,本文对“当前”统计模型进行了改进,通过对最大加速度的实时调整,从而克服了由固定加速度引起的收敛速度慢以及滤波发散问题,提高了跟踪的准确性。
标准Kalman滤波器对目标机动和运动模型选择不准的目标跟踪的适应能力较差,甚至导致跟踪失败,基于此,本文采用自适应Kalman滤波器结合“当前”统计模型,建立状态滤波算法,利用加速度和预测状态之间的关系,使最大加速度随着目标机动情况而进行自适应调整,进一步对加速度方差进行实时调整,从而使得系统噪声方差随着目标运动情况自适应变化,实现了目标的自适应跟踪。
考虑到跟踪的实时性,在建立目标的特征点表示模型时融合了目标的颜色和纹理信息,弥补了单纯利用目标灰度的匹配跟踪中对颜色不同车辆识别上存在的缺陷。经过对特征点的优化选择一方面减少了特征点的数量,提高了匹配速度；另一方面增加了特征点的信息量,提高了特征点在跟踪过程中的稳健性。
针对跟踪过程中存在遮挡及转弯变形问题,构建了一种基于特征点的图像跟踪算法,该算法在目标部分遮挡或旋转变形时都能够连续跟踪。根据匹配率和目标与背景的颜色差值建立了目标运动状态的判断机制及模板更新策略,解决了遮挡、转弯等情况下的目标跟踪问题。
最后,采用真实视频图像对本文算法进行了仿真实验,并获得了较好的跟踪效果,验证了本文算法的有效性。
Vision-based target tracking technology is a significant application of machine vision. It has attracted much attention in the research of Intelligent Transportation System(ITS). It is also one of the important research topics in the application of intelligent transportation monitoring. By tracking the moving vehicles we can extract the parameters of the traffic and detect the traffic issues automatically. During the vehicle tracking period, mutual covers of the vehicles, alteration of the light, influence of the complex background as well as the randomness of vehicle movement often happen, which affect the vehicle detection precision directly. How to track the vehicles steadily with complex background, mutual covers of the vehicles, alteration of the light and randomness of vehicle movement is the key aspect of this thesis.
In terms of the maneuverability of the vehicles, we build a "current" statistic model for moving targets. It has been widely accepted that the "current" statistic model performs well when tracking moving target but for targets with constant speed or targets with low speed, its tracking ability is poor. In order to solve this problem, this thesis presents an improved "current" statistic model. By adjusting the maximal acceleration, we overcome the drawback of slow convergence or divergence caused by constant acceleration and improve the tracking accuracy.
Traditional Kalman Filter is not suitable for moving target whose model is inaccurate. And it will even cause failure in target tracking. Take this phenomenon into account, in this thesis we combine adaptive Kalman Filter with "current" statistic model to build the state filter algorithm. Using the relation between acceleration and predictive state, the maximal acceleration as well as the variance of the maximal acceleration will adapt with the situation of the moving target. The variance of the system noise will also adapt with the situation of the moving target, so adaptive tracking of moving target can be achieved.
Take the real-time factor into account, we fuse the colour and the vein of the target when we use feature point to establish the model of the vehicles. So we can overcome the drawback when target grey is used as the sole feature to track the target, which is insensitive to vehicles with different colours. By optimizing feature points, we can not only reduce the number of the feature points and improve the match rate, but also increase the total information of the feature points and promote the stability of the tracking process.
We provide an image tracking algorithm based on feature point to solve the cover and deformation problem during the tracking process. This algorithm can track the target successfully even when the target is partially covered or has a revolvement deformation. By using matching rate as well as the colour difference between target and background, we can get the situation determination of the moving target and the update strategy of the background template. So we can solve the cover and deformation problem in target tracking process.
Finally, we use real video images to carry out some simulation experiments so as to test the robustness of our algorithm. Well performance is achieved and the effectiveness of our algorithm has been proved.
针对汽车运动过程的机动特点,建立了机动目标的“当前”统计模型,大量文献证明“当前”统计模型在跟踪机动目标时效果良好,但是其对匀速运动或弱机动目标跟踪时性能变差。针对这一问题,本文对“当前”统计模型进行了改进,通过对最大加速度的实时调整,从而克服了由固定加速度引起的收敛速度慢以及滤波发散问题,提高了跟踪的准确性。
标准Kalman滤波器对目标机动和运动模型选择不准的目标跟踪的适应能力较差,甚至导致跟踪失败,基于此,本文采用自适应Kalman滤波器结合“当前”统计模型,建立状态滤波算法,利用加速度和预测状态之间的关系,使最大加速度随着目标机动情况而进行自适应调整,进一步对加速度方差进行实时调整,从而使得系统噪声方差随着目标运动情况自适应变化,实现了目标的自适应跟踪。
考虑到跟踪的实时性,在建立目标的特征点表示模型时融合了目标的颜色和纹理信息,弥补了单纯利用目标灰度的匹配跟踪中对颜色不同车辆识别上存在的缺陷。经过对特征点的优化选择一方面减少了特征点的数量,提高了匹配速度；另一方面增加了特征点的信息量,提高了特征点在跟踪过程中的稳健性。
针对跟踪过程中存在遮挡及转弯变形问题,构建了一种基于特征点的图像跟踪算法,该算法在目标部分遮挡或旋转变形时都能够连续跟踪。根据匹配率和目标与背景的颜色差值建立了目标运动状态的判断机制及模板更新策略,解决了遮挡、转弯等情况下的目标跟踪问题。
最后,采用真实视频图像对本文算法进行了仿真实验,并获得了较好的跟踪效果,验证了本文算法的有效性。
Vision-based target tracking technology is a significant application of machine vision. It has attracted much attention in the research of Intelligent Transportation System(ITS). It is also one of the important research topics in the application of intelligent transportation monitoring. By tracking the moving vehicles we can extract the parameters of the traffic and detect the traffic issues automatically. During the vehicle tracking period, mutual covers of the vehicles, alteration of the light, influence of the complex background as well as the randomness of vehicle movement often happen, which affect the vehicle detection precision directly. How to track the vehicles steadily with complex background, mutual covers of the vehicles, alteration of the light and randomness of vehicle movement is the key aspect of this thesis.
In terms of the maneuverability of the vehicles, we build a "current" statistic model for moving targets. It has been widely accepted that the "current" statistic model performs well when tracking moving target but for targets with constant speed or targets with low speed, its tracking ability is poor. In order to solve this problem, this thesis presents an improved "current" statistic model. By adjusting the maximal acceleration, we overcome the drawback of slow convergence or divergence caused by constant acceleration and improve the tracking accuracy.
Traditional Kalman Filter is not suitable for moving target whose model is inaccurate. And it will even cause failure in target tracking. Take this phenomenon into account, in this thesis we combine adaptive Kalman Filter with "current" statistic model to build the state filter algorithm. Using the relation between acceleration and predictive state, the maximal acceleration as well as the variance of the maximal acceleration will adapt with the situation of the moving target. The variance of the system noise will also adapt with the situation of the moving target, so adaptive tracking of moving target can be achieved.
Take the real-time factor into account, we fuse the colour and the vein of the target when we use feature point to establish the model of the vehicles. So we can overcome the drawback when target grey is used as the sole feature to track the target, which is insensitive to vehicles with different colours. By optimizing feature points, we can not only reduce the number of the feature points and improve the match rate, but also increase the total information of the feature points and promote the stability of the tracking process.
We provide an image tracking algorithm based on feature point to solve the cover and deformation problem during the tracking process. This algorithm can track the target successfully even when the target is partially covered or has a revolvement deformation. By using matching rate as well as the colour difference between target and background, we can get the situation determination of the moving target and the update strategy of the background template. So we can solve the cover and deformation problem in target tracking process.
Finally, we use real video images to carry out some simulation experiments so as to test the robustness of our algorithm. Well performance is achieved and the effectiveness of our algorithm has been proved.
引文
[1]张广军.机器视觉[M].北京：科学出版社.2008.4.
[2]徐友春,李克强,连小珉.智能车辆机器视觉发展近况[J].汽车工程.2003,Vol.25No.5：438-443.
[3]刘会杰,王学孝.计算机视觉的理论框架研究[J].计算机科学.1999,vol.26No.3：34-37.
[4]崔岩.基于机器视觉的运动目标轨迹跟踪方法研究[D].[硕士学位论文].沈阳：东北大学.2005年7月.
[5]段峰,王耀南,雷晓峰等.机器视觉技术及其应用综述[J].自动化博览：2002：59-61.
[6]沈中杰,王武宏,钟永刚.智能交通系统中基于机器视觉的数字车辆控制[J].汽车工程.2003(Vol.25)No.5：428-433.
[7]Courtney R.L.A broad view of ITS standards in the US. Intelligent Transportation System.1997. ITSC 97. IEEE Conference on 9-12 Nov.1997 Page(s):529-536.
[8]史其信.21世纪的智能交通系统ITS(上).交通世界[J].2001,第2-3期：62-65.
[9]史其信.21世纪的智能交通系统ITS(上).交通世界[J].2001,第4期：30-33.
[10]侯志强,韩崇昭.视觉跟踪技术综述[J].自动化学报,2006,第32卷,第4期：603-617.
[11]张达旭.运动目标跟踪算法研究[D].[硕士学位论文].沈阳：东北大学,2006.2.
[12]高恩克.目标跟踪的算法研究[D].[硕士学位论文].西安：西安电子科技大学2007.1.
[13]王圣男,郁梅等.智能交通系统中基于视频图像处理的车辆检测与跟踪方法综述[J].计算机应用研究.2005,第9期：9-14.
[14]Rabiner L, Juang B. An introduction to hidden Markov models [J]. IEEE on Signal Processing.1986,3(1):4-16.
[15]瓦普尼克,张学工.统计学习理论的本质[M].北京：清华大学出版社.2000.
[16]Arulampalam MS, Maskell S, Gordon N, et al. A tutorial on particle filters for online nonlinear/non-GaussianBayesian tracking [J]. IEEE Transactions on Signal Processing,2002,50(2):174-188.
[17]中国科学院自动化研究所模式识别国家重点实验室视觉监控研究组.智能视觉监控技术的研究背景现状及展望.摘自今日机电网.2006年.
[18]许录平.数字图像处理[M].北京：科学出版社.2007.10.
[19]赵荣椿编著.数字图像处理导论[M].西安：西北工业大学出版社.2000.
[20]Sinisa Todorovic,Narendra Ahuja.Region-Based Hierarchical Image Matching[J].Int J Comput Vis (2008) 78:47-66.
[21]李安平.复杂环境下的视频目标跟踪算法研究[D].[博士学位论文].上海交通大学.2006.10.
[22]Kass M, Witkin A, Terzopoulous D. Snakes:active contour models. In:Brady IM, Rosenfield A eds[C]. Proceedings of the 1st International Conference on Computer Vision. London:IEEE Computer Society Press.1987:259-268.
[23]Benhimane S, Ladikos A, Lepetit V, Navab N. Linear and Quadratic Subsets for Template-Based Tracking[C]. Computer Vision and Pattern Recognition,2007. CVPR'07. IEEE Conference on 17-22 June 2007:1-6.
[24]Sungho Kim,In So Kweon.Automatic model-based 3D object recognition by combining feature matching with tracking[J].Machine Vision and Applications (2005) 16(5):267-272.
[25]Tim K. Marks, John Hershey, J. Cooper Roddey, Javier R. Movellan.3D Tracking of Morphable ObjectsUsing Conditionally Gaussian Nonlinear Filters[J].Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW'04).
[26]GORDON WELLS and CARME TORRAS.Assessing Image Features for Vision-Based Robot Positioning[J].Journal of Intelligent and Robotic Systems 30(2001):95-118.
[27]Tun Hu,Erqi Liu,Jie Yang.Multi-feature Based Ensemble Classification and Regression Tree (ECART) for Target Tracking in Infrared Imagery[J].J Infrared Milli Terahz Waves (2009) 30:484-495.
[28]王伟.基于视觉的运动汽车跟踪技术研究[D].[硕士学位论文].兰州理工大学.2009.6.
[29]Singer R A. Estimating optimal tracking filter performance for manned maneu-vering targets[C].IEEE Transactions on Aweospace and Electronic Systems, 1970,6(4):473～483.
[30]Yan Peng, Hongbin Jin.A maneuvering target tracking algorithm based on UKF-Singer[C].System Simulation and Scientific Computing,2008. ICSC 2008. Asia Simulation Conference-7th International Conference on 10-12 Oct.2008 Page(s): 924～926.
[31]Kendrick J D, Maybeck P S, Reid JG.Estimation of aircraft target motion using orientation measurements[C]. IEEE Transaction on Aerospace and Electronic Systems.1981,17(3):254～259.
[32]Mahapatra P R, Kishore M. Mixed coordinate tracking generalized maneuvering targets using acceleration Jerk Models. IEEE Trans.Aerospace and Electronic SystemsMar.2000:992～1000.
[33]Kishore M, Mahapatra P R.A Jerk model for tracking highly maneuvering targets[C].IEEE Trans.Aerospace and Electronic Systems,Apr.1997:1094～1105.
[34]乔向东,王宝树.高度机动目标的‘当前’统计Jerk模型[J].系统工程与电子技术.2000,24(6)：53-56.
[35]周宏仁.机动目标当前统计模型与自适应跟踪算法[J].航空学报.1983,4(1)：73-86.
[36]罗笑冰,王宏强,黎湘,庄钊文.非线性“当前”统计模型及自适应跟踪算法[J].系统工程与电子技术.2008.3,Vol.30,No.3：397-400.
[36]马安国,罗德林,沈春林,龚华军.基于“当前l”统计模型的目标状态估计[J].控制理论与应用.2006,Vol.25,No.7：5-8.
[37]姚洪利.用改进的“当前”统计模型对机动目标快速跟踪[J].情报指挥控制系统与仿真技术.2004.12,Vol.26,No.6：38-43.
[38]白建华.基于特征点的运动车辆目标检测与跟踪[D].[硕士学位论文].厦门大学.2006.5：17-18.
[39]D.P.Huttenlocker,J.J.Nob,W.J.Rucklidge."Tracking non-rigid objects in complex scenes".The 4th International Conference on Computer Vision,Berlin Germany, May 1993.
[40]K.Nummiaro,E. Koller-Meier,L.Van Gool. " An adaptive color-based particle filter".Image and Vision Computing,2003,21:99-110.
[41]Z. Zhnag, R. Deriche,O Fuageras. A Robust Technique for Matching Two Uncalibrated Images Through the Recovery of the Unknown Epipolar Geometry [J]. Artifieial Intelligence.1995,78(1-2):87-119.
[42]Pan Jun-Jun,Zhang Yan-ning. Correspongding Points, Matching Based on Position Similarity.IEEE proceedings of the Computer Graphics,Imaging and Vsion:New Trends.2005:5154-5158.
[43]张志佳.图像特征提取及目标匹配的理论和方法研究[D].[博士学位论文].中国科学院沈阳自动化研究所.2006.1.
[44]JOAQUIN FERRUZ,ANIBAL OLLERO.Real-Time Feature Matching in Image Sequences for Non-Structured Environments. Applications to Vehicle Guidance.Journal of Intelligent and Robotic Systems 2000,28:85-123.
[45]刘丽娜.图像特征点匹配算法研究[D].[硕士学位论文].南京理工大学.2006.
[46]屈志毅,黄鹤鸣,孔令旺.利用Mahalanobis距离进行人脸表情的识别[J].兰州大学学报(自然科学版).2005,vol.41,No.6.
[47]张绍璞.Mahalanobis距离在多元随机变量线性变换中的定理[J].天津科技大学学报.2008.6,vol.23,No.2.
[48]杨万海.多传感器数据融合及其应用[M].西安电子科技大学出版社.2004.
[49]胡士强,敬忠良.粒子滤波算法综述[J].控制与决策.2005.4,Vol20,No.4：361-365.
[50]H. P. Moravec, Towards Automatic Visual Obstacle Avoidance[J]. Proc.5th International Joint Conference on Artificial Intelligence, pp.584,1977.
[51]刘达.快速特征点匹配技术的研究和实现[D].[硕士学位论文].燕山大学.2006.
[52]张春森.基于点特征匹配的SUSAN、Harris算子比较.测绘信息网www.othermap.com.
[53]张春美,龚志辉,黄艳.几种特征点提取算法的性能评估及改进[J].测绘科学技术学报.2008,Vol.25,No.3.
[54]C. Harris, M. Stephens.A Combined Corner and Edge Detector[J]. Proc. Alvey Vision Conf. Univ. Mancheste.1988:147-151.
[55]张广军.机器视觉[M].北京：科学出版社.2008.4：50-52.
[56]刘荆桥,戴光明.角点检测在车辆特征提取中的应用[J].软件导刊.Apr.2008.Vol.7,No.4：91-92.
[57]Gevers K, Smeulders AWM.Color-based object recognition[J].Pattern Recognition.1999,32:453-464.
[58]高健,黄心汉,彭刚,王敏,吴祖玉.基于彩色的SIFT特征点提取与匹配[J].计算机工程与应用.2007,43(34)：10-15.
[59]常向魁.运动视觉中目标的精确提取与跟踪技术[D].[硕士学位论文].河南大学.2007.
[60]王宇,程耀瑜.基于Kalman滤波原理的运动目标跟踪[J].信息技术.2008.10：48-51.
[61]赵其杰,屠大维,高健,黄志华.基于Kalman滤波的视觉预测目标跟踪及其应用[J].光学精密工程.2008.Vol.16,No.5：937-941.
[62]虞旦,韦巍,张远辉.一种基于卡尔曼预测的动态目标跟踪算法研究[J].光电工程.2009.Vol.36,No.1：52-56.
[63]刘海燕,赵宗贵,刘熹.“当前”统计模型机动目标跟踪的改进算法[J].军事运筹与系统工程.2007.Vol.21 No.1：51-54.
[64]段战胜,韩崇昭,党宏社.最大加速度未知的“当前”统计模型机动目标跟踪[J].计算机工程与应用.2003.29：19-22.
[65]刁联旺,杨静宇.一种改进的机动目标“当前”统计模型的描述[J].兵工学报.2005,Vol.26 No.6：825-828.
[66]罗笑冰,王宏强,黎湘,庄钊文.非线性“当前”统计模型及自适应跟踪算法[J].系 统工程与电子技术.2008,Vol.30,No.3：397-403.
[67]方青,梅晓春,张育平.用于机动目标跟踪的Kalman滤波器的设计[J].雷达科学与技术.2006,Vol.4 No.1：50-55.
[2]徐友春,李克强,连小珉.智能车辆机器视觉发展近况[J].汽车工程.2003,Vol.25No.5：438-443.
[3]刘会杰,王学孝.计算机视觉的理论框架研究[J].计算机科学.1999,vol.26No.3：34-37.
[4]崔岩.基于机器视觉的运动目标轨迹跟踪方法研究[D].[硕士学位论文].沈阳：东北大学.2005年7月.
[5]段峰,王耀南,雷晓峰等.机器视觉技术及其应用综述[J].自动化博览：2002：59-61.
[6]沈中杰,王武宏,钟永刚.智能交通系统中基于机器视觉的数字车辆控制[J].汽车工程.2003(Vol.25)No.5：428-433.
[7]Courtney R.L.A broad view of ITS standards in the US. Intelligent Transportation System.1997. ITSC 97. IEEE Conference on 9-12 Nov.1997 Page(s):529-536.
[8]史其信.21世纪的智能交通系统ITS(上).交通世界[J].2001,第2-3期：62-65.
[9]史其信.21世纪的智能交通系统ITS(上).交通世界[J].2001,第4期：30-33.
[10]侯志强,韩崇昭.视觉跟踪技术综述[J].自动化学报,2006,第32卷,第4期：603-617.
[11]张达旭.运动目标跟踪算法研究[D].[硕士学位论文].沈阳：东北大学,2006.2.
[12]高恩克.目标跟踪的算法研究[D].[硕士学位论文].西安：西安电子科技大学2007.1.
[13]王圣男,郁梅等.智能交通系统中基于视频图像处理的车辆检测与跟踪方法综述[J].计算机应用研究.2005,第9期：9-14.
[14]Rabiner L, Juang B. An introduction to hidden Markov models [J]. IEEE on Signal Processing.1986,3(1):4-16.
[15]瓦普尼克,张学工.统计学习理论的本质[M].北京：清华大学出版社.2000.
[16]Arulampalam MS, Maskell S, Gordon N, et al. A tutorial on particle filters for online nonlinear/non-GaussianBayesian tracking [J]. IEEE Transactions on Signal Processing,2002,50(2):174-188.
[17]中国科学院自动化研究所模式识别国家重点实验室视觉监控研究组.智能视觉监控技术的研究背景现状及展望.摘自今日机电网.2006年.
[18]许录平.数字图像处理[M].北京：科学出版社.2007.10.
[19]赵荣椿编著.数字图像处理导论[M].西安：西北工业大学出版社.2000.
[20]Sinisa Todorovic,Narendra Ahuja.Region-Based Hierarchical Image Matching[J].Int J Comput Vis (2008) 78:47-66.
[21]李安平.复杂环境下的视频目标跟踪算法研究[D].[博士学位论文].上海交通大学.2006.10.
[22]Kass M, Witkin A, Terzopoulous D. Snakes:active contour models. In:Brady IM, Rosenfield A eds[C]. Proceedings of the 1st International Conference on Computer Vision. London:IEEE Computer Society Press.1987:259-268.
[23]Benhimane S, Ladikos A, Lepetit V, Navab N. Linear and Quadratic Subsets for Template-Based Tracking[C]. Computer Vision and Pattern Recognition,2007. CVPR'07. IEEE Conference on 17-22 June 2007:1-6.
[24]Sungho Kim,In So Kweon.Automatic model-based 3D object recognition by combining feature matching with tracking[J].Machine Vision and Applications (2005) 16(5):267-272.
[25]Tim K. Marks, John Hershey, J. Cooper Roddey, Javier R. Movellan.3D Tracking of Morphable ObjectsUsing Conditionally Gaussian Nonlinear Filters[J].Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW'04).
[26]GORDON WELLS and CARME TORRAS.Assessing Image Features for Vision-Based Robot Positioning[J].Journal of Intelligent and Robotic Systems 30(2001):95-118.
[27]Tun Hu,Erqi Liu,Jie Yang.Multi-feature Based Ensemble Classification and Regression Tree (ECART) for Target Tracking in Infrared Imagery[J].J Infrared Milli Terahz Waves (2009) 30:484-495.
[28]王伟.基于视觉的运动汽车跟踪技术研究[D].[硕士学位论文].兰州理工大学.2009.6.
[29]Singer R A. Estimating optimal tracking filter performance for manned maneu-vering targets[C].IEEE Transactions on Aweospace and Electronic Systems, 1970,6(4):473～483.
[30]Yan Peng, Hongbin Jin.A maneuvering target tracking algorithm based on UKF-Singer[C].System Simulation and Scientific Computing,2008. ICSC 2008. Asia Simulation Conference-7th International Conference on 10-12 Oct.2008 Page(s): 924～926.
[31]Kendrick J D, Maybeck P S, Reid JG.Estimation of aircraft target motion using orientation measurements[C]. IEEE Transaction on Aerospace and Electronic Systems.1981,17(3):254～259.
[32]Mahapatra P R, Kishore M. Mixed coordinate tracking generalized maneuvering targets using acceleration Jerk Models. IEEE Trans.Aerospace and Electronic SystemsMar.2000:992～1000.
[33]Kishore M, Mahapatra P R.A Jerk model for tracking highly maneuvering targets[C].IEEE Trans.Aerospace and Electronic Systems,Apr.1997:1094～1105.
[34]乔向东,王宝树.高度机动目标的‘当前’统计Jerk模型[J].系统工程与电子技术.2000,24(6)：53-56.
[35]周宏仁.机动目标当前统计模型与自适应跟踪算法[J].航空学报.1983,4(1)：73-86.
[36]罗笑冰,王宏强,黎湘,庄钊文.非线性“当前”统计模型及自适应跟踪算法[J].系统工程与电子技术.2008.3,Vol.30,No.3：397-400.
[36]马安国,罗德林,沈春林,龚华军.基于“当前l”统计模型的目标状态估计[J].控制理论与应用.2006,Vol.25,No.7：5-8.
[37]姚洪利.用改进的“当前”统计模型对机动目标快速跟踪[J].情报指挥控制系统与仿真技术.2004.12,Vol.26,No.6：38-43.
[38]白建华.基于特征点的运动车辆目标检测与跟踪[D].[硕士学位论文].厦门大学.2006.5：17-18.
[39]D.P.Huttenlocker,J.J.Nob,W.J.Rucklidge."Tracking non-rigid objects in complex scenes".The 4th International Conference on Computer Vision,Berlin Germany, May 1993.
[40]K.Nummiaro,E. Koller-Meier,L.Van Gool. " An adaptive color-based particle filter".Image and Vision Computing,2003,21:99-110.
[41]Z. Zhnag, R. Deriche,O Fuageras. A Robust Technique for Matching Two Uncalibrated Images Through the Recovery of the Unknown Epipolar Geometry [J]. Artifieial Intelligence.1995,78(1-2):87-119.
[42]Pan Jun-Jun,Zhang Yan-ning. Correspongding Points, Matching Based on Position Similarity.IEEE proceedings of the Computer Graphics,Imaging and Vsion:New Trends.2005:5154-5158.
[43]张志佳.图像特征提取及目标匹配的理论和方法研究[D].[博士学位论文].中国科学院沈阳自动化研究所.2006.1.
[44]JOAQUIN FERRUZ,ANIBAL OLLERO.Real-Time Feature Matching in Image Sequences for Non-Structured Environments. Applications to Vehicle Guidance.Journal of Intelligent and Robotic Systems 2000,28:85-123.
[45]刘丽娜.图像特征点匹配算法研究[D].[硕士学位论文].南京理工大学.2006.
[46]屈志毅,黄鹤鸣,孔令旺.利用Mahalanobis距离进行人脸表情的识别[J].兰州大学学报(自然科学版).2005,vol.41,No.6.
[47]张绍璞.Mahalanobis距离在多元随机变量线性变换中的定理[J].天津科技大学学报.2008.6,vol.23,No.2.
[48]杨万海.多传感器数据融合及其应用[M].西安电子科技大学出版社.2004.
[49]胡士强,敬忠良.粒子滤波算法综述[J].控制与决策.2005.4,Vol20,No.4：361-365.
[50]H. P. Moravec, Towards Automatic Visual Obstacle Avoidance[J]. Proc.5th International Joint Conference on Artificial Intelligence, pp.584,1977.
[51]刘达.快速特征点匹配技术的研究和实现[D].[硕士学位论文].燕山大学.2006.
[52]张春森.基于点特征匹配的SUSAN、Harris算子比较.测绘信息网www.othermap.com.
[53]张春美,龚志辉,黄艳.几种特征点提取算法的性能评估及改进[J].测绘科学技术学报.2008,Vol.25,No.3.
[54]C. Harris, M. Stephens.A Combined Corner and Edge Detector[J]. Proc. Alvey Vision Conf. Univ. Mancheste.1988:147-151.
[55]张广军.机器视觉[M].北京：科学出版社.2008.4：50-52.
[56]刘荆桥,戴光明.角点检测在车辆特征提取中的应用[J].软件导刊.Apr.2008.Vol.7,No.4：91-92.
[57]Gevers K, Smeulders AWM.Color-based object recognition[J].Pattern Recognition.1999,32:453-464.
[58]高健,黄心汉,彭刚,王敏,吴祖玉.基于彩色的SIFT特征点提取与匹配[J].计算机工程与应用.2007,43(34)：10-15.
[59]常向魁.运动视觉中目标的精确提取与跟踪技术[D].[硕士学位论文].河南大学.2007.
[60]王宇,程耀瑜.基于Kalman滤波原理的运动目标跟踪[J].信息技术.2008.10：48-51.
[61]赵其杰,屠大维,高健,黄志华.基于Kalman滤波的视觉预测目标跟踪及其应用[J].光学精密工程.2008.Vol.16,No.5：937-941.
[62]虞旦,韦巍,张远辉.一种基于卡尔曼预测的动态目标跟踪算法研究[J].光电工程.2009.Vol.36,No.1：52-56.
[63]刘海燕,赵宗贵,刘熹.“当前”统计模型机动目标跟踪的改进算法[J].军事运筹与系统工程.2007.Vol.21 No.1：51-54.
[64]段战胜,韩崇昭,党宏社.最大加速度未知的“当前”统计模型机动目标跟踪[J].计算机工程与应用.2003.29：19-22.
[65]刁联旺,杨静宇.一种改进的机动目标“当前”统计模型的描述[J].兵工学报.2005,Vol.26 No.6：825-828.
[66]罗笑冰,王宏强,黎湘,庄钊文.非线性“当前”统计模型及自适应跟踪算法[J].系 统工程与电子技术.2008,Vol.30,No.3：397-403.
[67]方青,梅晓春,张育平.用于机动目标跟踪的Kalman滤波器的设计[J].雷达科学与技术.2006,Vol.4 No.1：50-55.